Design-for-manufacturing (DFM) is a developmental practice emphasizing manufacturing issues throughout the product development process. Successful DFM results in lower production cost without sacrificing product quality.
The concept of DFM has been widely adopted in the manufacturing industry, particularly in the manufacturing of integrated circuits. The existing DFM systems, however, are often collections of different systems from different vendors, for example, electrical data analysis (EDA), third party intellectual property (IP) cell platforms, hot spot management systems, failure report systems, and the like. These heterogeneous systems not only cause the reduction in the efficiency of the DFM systems, but also create barriers for the information flow. The identified issues and created actions stored in one system cannot be ported to other related systems. For example, a problem identified in 90 nm technology needs to be manually checked and, if found, fixed in 65 nm technology. There is no mechanism for tracking and ensuring the execution of the problem identification and solution.
Further, DFM models and IP library platforms themselves need to be updated from time to time, such modifications make it necessary for further DFM re-checks using the modified DFM models. In existing DFM systems, the needed DFM re-checks are only identified and initiated manually. Therefore, identifying products related to the modified DFM models and IP libraries is itself a big task, and tracking the progress and the results of the DFM re-checks is an even more daunting one. Accordingly, a DFM system for solving the above-discussed problems is needed.